Power grid topology abnormality alarm data processing method and system based on multi-converter

Abstract

The present invention relates to the field of microgrid abnormal alarm, and in particular to a method and system for processing data of abnormal alarm of power grid topology based on multiple converters. The method comprises the following steps: identifying microgrid converter nodes based on microgrid converter clusters, and obtaining microgrid historical converter node state parameters; performing deep characterization association mining on microgrid historical converter node state parameters, and constructing node dynamic monitoring feature maps; performing node topology connection logic mining on microgrid converter nodes one by one, and performing global topology structure fitting, and constructing a microgrid global topology structure model; using node dynamic monitoring feature maps to perform quantitative calculation of feature trajectory deviation on the microgrid global topology structure model, thereby obtaining abnormal dynamic state feature trajectory; performing normalized node filtering based on abnormal dynamic state feature trajectory, thereby obtaining abnormal topology feature nodes. The present invention realizes efficient and accurate microgrid topology abnormal alarm.

Description

Technical Field

[0001] The present invention relates to the field of microgrid abnormality alarm, and in particular to a multi-converter-based power grid topology abnormality alarm data processing method and system. Background Art

[0002] With the continuous advancement of smart microgrid construction, the complexity and dynamics of microgrid systems are constantly increasing. Microgrids have gradually evolved from traditional single topology structures to complex networks containing a large number of distributed generation units, energy storage systems, loads and other nodes. In practical applications, due to the continuous updating and iteration of microgrids, the number of converters in the region has increased significantly. The access of converters may cause the overall topology of the microgrid to become more complex, resulting in complex fault diagnosis, and the defects and failure rate of the power grid will also increase. The topology of this microgrid system is becoming more and more dynamic and changeable, which brings new challenges to the stable operation of the microgrid. Abnormal changes in microgrid topology are caused by natural disasters, man-made accidents or equipment failures. Once they occur, they will lead to serious consequences such as power supply interruption, equipment damage, and safety hazards. Therefore, timely and accurate detection and diagnosis of abnormal changes in microgrid topology is crucial to ensure the safe and stable operation of microgrids. Traditional microgrid topology monitoring methods mainly rely on manual inspections, fault signal detection, etc., which have problems of low efficiency and low diagnostic accuracy. With the rapid development of microgrid converter technology, converter clusters monitor key parameters such as voltage, current, and power factor of each node in the microgrid in real time, providing new technical support for the intelligent detection of microgrid topology anomalies. Based on this, it is urgent to study an intelligent microgrid topology anomaly alarm method. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a method and system for processing power grid topology abnormality alarm data based on multiple converters to solve at least one of the above technical problems.

[0004] To achieve the above object, the present invention provides a method for processing abnormal alarm data of power grid topology based on multi-converter, comprising the following steps:

[0005] Step S1: Identify microgrid converter nodes based on microgrid converter clusters and obtain microgrid historical converter node state parameters; perform deep characterization correlation mining on microgrid historical converter node state parameters to construct node dynamic monitoring feature maps;

[0006] Step S2: mining the topological connection logic of each microgrid converter node one by one, fitting the global topological structure, and constructing a microgrid global topological structure model;

[0007] Step S3: Quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory;

[0008] Step S4: Perform normalized node filtering based on the abnormal dynamic state feature trajectory, and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes;

[0009] Step S5: performing a topological cascade effect analysis on the abnormal topological feature nodes to generate abnormal node topological cascade effect data; performing a topological local fracture analysis based on the abnormal node topological cascade effect data to obtain an abnormal topological local fracture area;

[0010] Step S6: Make personalized alarm decisions for each node in the abnormal topology local fracture area, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

[0011] The present invention identifies microgrid converter nodes and obtains historical state parameters, establishes a historical data foundation for microgrid nodes, provides a basis for subsequent analysis, conducts deep characterization association mining and constructs node dynamic monitoring feature maps to discover association patterns between nodes, lays a foundation for anomaly detection, mines the topological connection logic of each node and fits the global topological structure to reveal the connection relationship between microgrid nodes, constructs a microgrid global topological structure model, understands the structure of the entire microgrid system, and provides a global perspective for subsequent anomaly detection. The node dynamic monitoring feature map is used to perform quantitative calculation of feature trajectory deviation on the global topological structure model, discovers abnormal dynamic state feature trajectories, helps identify potential problems, and obtains abnormal dynamic state feature trajectories to detect microgrid problems in advance. Abnormal situations in the network are detected by filtering regular nodes based on the abnormal dynamic state feature trajectory and accurately locating abnormal nodes to accurately find abnormal nodes, narrow the scope of the problem, and obtain abnormal topological feature nodes to help operation and maintenance personnel quickly locate problems and reduce troubleshooting time. Topological cascade effect analysis is performed on abnormal topological feature nodes to understand the impact of abnormal nodes on the entire microgrid system, help assess risks and prioritize nodes, and perform topological local fracture analysis based on the topological cascade effect data of abnormal nodes to determine the abnormal topological local fracture area, further refine the scope of the problem, and make personalized alarm decisions for the abnormal topological local fracture area. A topological abnormality alarm optimization model is constructed to achieve more accurate alarms and processing, and improve the stability and security of the microgrid system.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: identifying microgrid converter nodes based on the microgrid converter cluster, and obtaining microgrid historical converter node state parameters;

[0014] Step S12: performing multi-time point voltage load calculation on microgrid historical converter node state parameters to generate converter node voltage load data;

[0015] Step S13: performing time series load trend analysis on the converter node voltage load data to obtain a node load trend curve;

[0016] Step S14: performing temperature state identification on microgrid historical converter node state parameters to generate node temperature characteristics;

[0017] Step S15: performing temperature rise variation trend evolution on the node temperature characteristics, thereby obtaining node temperature rise variation characteristic data;

[0018] Step S16: Based on the node temperature rise change characteristic data, deep characterization and correlation mining are performed on the node load trend curve to construct a node dynamic monitoring characteristic map.

[0019] The present invention identifies microgrid converter nodes and obtains historical status parameters, establishes a complete node status record for the microgrid system, provides basic data for subsequent analysis, obtains microgrid historical converter node status parameters to understand the node operation status, provides necessary information for further analysis of node load and temperature, performs multi-time point voltage load calculation to generate converter node voltage load data, helps evaluate the node power consumption and load status, obtains converter node voltage load data to monitor the node power consumption, provides data support for load analysis and trend prediction, performs time series load trend analysis on converter node voltage load data to reveal the node load change trend, helps evaluate the node load status, obtains the node load trend curve to understand the node load change law, and provides a reference basis for abnormal detection and load management , the temperature status of the microgrid historical converter node state parameters is identified to generate the temperature characteristics of the node, which helps to monitor the temperature of the node. The node temperature characteristics are obtained to understand the heat distribution of the node, which provides a basis for the temperature rise trend analysis of the node. The temperature rise change trend evolution analysis of the node temperature characteristics is performed to obtain the node temperature rise change characteristic data, which helps to evaluate the temperature change of the node. The node temperature rise change characteristic data is obtained to monitor the temperature change trend of the node, which provides a basis for node status evaluation and prediction. Based on the node temperature rise change characteristic data, the node load trend curve is deeply characterized and associated with mining to construct a node dynamic monitoring feature map. Considering the node load and temperature information comprehensively, the node dynamic measurement feature map is constructed to discover the correlation pattern between nodes, which provides more comprehensive node status information for anomaly detection and topology anomaly alarm systems.

[0020] Preferably, the specific steps of step S16 are:

[0021] Based on the node temperature rise variation characteristic data, a temperature-load correlation analysis is performed on the node load trend curve to generate temperature-load correlation data;

[0022] Perform multi-dimensional feature space mapping on the converter node voltage load data and node temperature rise change feature data to generate a node feature multi-dimensional space model;

[0023] Based on the temperature-load correlation data, the node dynamic feature drift analysis is performed on the node feature multidimensional space model to generate the node dynamic drift trajectory;

[0024] Conduct deep characterization and correlation mining on the dynamic drift trajectory of nodes and construct a node dynamic monitoring feature map.

[0025] The present invention reveals the correlation between node temperature change and load trend through temperature-load correlation analysis, helps to understand the temperature change law of nodes under different load conditions, generates temperature-load correlation data to establish a correlation model between node temperature and load, provides a basis for subsequent dynamic feature drift analysis, maps converter node voltage load data and node temperature rise change feature data to a multidimensional feature space, comprehensively considers the impact of different features on node status, provides a more comprehensive node feature representation, and provides data support for subsequent dynamic feature drift analysis and correlation mining. By analyzing the temperature-load correlation data of the node feature multidimensional space model, the node dynamic feature drift is detected, that is, the trend of node status changing over time, and the change mode and trend of the node status are identified, providing a more accurate basis for abnormal detection and early warning. Based on the node dynamic drift trajectory, deep characterization correlation mining is performed to construct a node dynamic monitoring feature map, comprehensively considers the multi-dimensional information of the node characteristics, constructs a node dynamic monitoring feature map to discover the correlation between node states, and provides a more comprehensive node state monitoring and analysis for the topology abnormality alarm system.

[0026] Preferably, the specific steps of step S2 are:

[0027] Step S21: Calculate the spatial position of the microgrid converter node to obtain the node spatial position coordinates;

[0028] Step S22: performing node spatial distribution analysis based on the node spatial position coordinates to obtain node spatial distribution data;

[0029] Step S23: performing microgrid topology analysis on the node spatial distribution data to obtain a microgrid topology diagram;

[0030] Step S24: mining the topological connection logic of each microgrid converter node one by one, and extracting the topological association logic between nodes;

[0031] Step S25: performing global topological structure fitting on the microgrid topological structure diagram according to the topological association logic between nodes, and constructing a global topological structure model of the microgrid.

[0032] The present invention calculates the spatial position of microgrid converter nodes, obtains the accurate position coordinates of each node in the microgrid, obtains the node spatial position coordinates to establish the spatial relationship between nodes, and provides basic data for subsequent spatial distribution analysis and topological structure analysis. The node spatial distribution analysis is performed based on the node spatial position coordinates to understand the spatial layout of each node in the microgrid, including the distance and position relationship between the nodes, and the node spatial distribution data is obtained to evaluate the spatial distribution characteristics between the nodes, providing a reference for the spatial perspective of the microgrid topological structure analysis. The microgrid topological structure analysis of the node spatial distribution data reveals the topological connection relationship between the nodes in the microgrid system, and forms a microgrid topological structure. The topological structure diagram of the network is obtained to understand the structural organization of the microgrid system, which provides a basis for subsequent logic mining and global topological structure model construction. By mining the topological connection logic of each node, the topological association logic between nodes is extracted, including the connection mode and dependency relationship between nodes, so as to deeply understand the relationship between nodes and provide logical support for the global fitting and modeling of the microgrid topological structure. According to the topological association logic between nodes, the global topological structure of the microgrid topological structure diagram is fitted, and the global topological structure model of the microgrid is constructed. The association relationship between each node in the microgrid system is comprehensively considered to provide global perspective support for the detection and alarm of microgrid topological anomalies.

[0033] Preferably, step S3 specifically comprises the following steps:

[0034] Step S31: performing node dynamic state characteristic monitoring simulation on the microgrid global topology structure model to extract node dynamic state characteristic data;

[0035] Step S32: performing time series trajectory evolution on the node dynamic state feature data to generate a node dynamic state evolution trajectory;

[0036] Step S33: using the node dynamic monitoring characteristic map to perform quantitative calculation of characteristic trajectory deviation on the node dynamic state evolution trajectory to obtain dynamic characteristic trajectory deviation data;

[0037] Step S34: performing abnormal trajectory feature analysis on the dynamic feature trajectory deviation data, thereby obtaining an abnormal dynamic state feature trajectory.

[0038] The present invention simulates node dynamic state characteristic monitoring on a global topological structure model of a microgrid, simulates the characteristic changes of nodes under different states, extracts node dynamic state characteristic data, extracts node dynamic state characteristic data to understand the changes of node states over time, provides a data basis for subsequent time series trajectory evolution and anomaly detection, performs time series trajectory evolution on node dynamic state characteristic data to generate node dynamic state evolution trajectory, displays the change trend of node states over time, observes the historical evolution process of node states, provides time series data support for subsequent characteristic trajectory deviation analysis, uses node dynamic monitoring characteristic maps to perform quantitative calculation of characteristic trajectory deviation on node dynamic state evolution trajectory, quantifies the degree of deviation between node dynamic states and expected states, identifies abnormal situations of node dynamic state changes, provides quantitative basis for abnormal trajectory characteristic analysis, performs abnormal trajectory characteristic analysis on dynamic characteristic trajectory deviation data to identify and analyze abnormal dynamic state characteristic trajectories, that is, abnormal situations occurring in node state changes, identifies abnormal dynamic state characteristic trajectories to monitor and warn of microgrid topology abnormalities in real time, and improves the safety and stability of the microgrid system.

[0039] Preferably, the specific steps of step S4 are:

[0040] Step S41: dynamically positioning the global topology structure model of the microgrid based on the abnormal dynamic state characteristic trajectory to obtain abnormal trajectory positioning data;

[0041] Step S42: performing node identification on the abnormal trajectory positioning data, and marking all converter nodes on the abnormal trajectory;

[0042] Step S43: performing normalized node filtering on all converter nodes on the abnormal trajectory and accurately locating the abnormal nodes, thereby obtaining abnormal topological feature nodes.

[0043] The present invention locates the dynamic operation trajectory of the global topological structure model of the microgrid based on the abnormal dynamic state characteristic trajectory, accurately locates the specific position of the abnormal trajectory in the microgrid, obtains abnormal trajectory positioning data, quickly locates the position where the abnormality occurs in the microgrid system, and provides a basis for subsequent abnormal processing and repair. The abnormal trajectory positioning data is used to identify nodes on the trajectory, mark all converter nodes on the abnormal trajectory, identify all nodes passed by the abnormal trajectory, determine the microgrid node range corresponding to the abnormal trajectory, provide information support for the next step of node filtering and abnormal node positioning, perform regular node filtering on all converter nodes on the abnormal trajectory, accurately locate the abnormal nodes, accurately identify and locate the abnormal topological characteristic nodes, that is, the specific nodes where the abnormality occurs, deeply analyze the cause of the abnormality, quickly respond to the abnormal situation, and improve the fault location and processing efficiency of the microgrid system.

[0044] Preferably, the specific steps of step S5 are:

[0045] Step S51: quantifying the topological stability impact of abnormal topological characteristic nodes to generate abnormal node topological impact values;

[0046] Step S52: performing a topological cascade effect analysis on the abnormal node topological impact value to generate abnormal node topological cascade effect data;

[0047] Step S53: performing a numerical simulation of topological anomaly propagation on the global topological structure model of the microgrid according to the abnormal node topological cascade effect data, thereby generating an abnormal node propagation topological path;

[0048] Step S54: Perform a topological local fracture analysis on the abnormal node propagation topological path to obtain an abnormal topological local fracture area.

[0049] The present invention quantifies the impact of topological stability on abnormal topological characteristic nodes, quantifies the degree of influence of abnormal nodes on the stability of microgrid systems, evaluates the degree of topological structure changes caused by abnormal nodes, provides quantitative basis for subsequent cascade effect analysis, performs topological cascade effect analysis on the topological impact value of abnormal nodes, analyzes how the influence of abnormal nodes spreads to surrounding nodes, generates abnormal node topological cascade effect data to understand the degree of influence of abnormal node failures on surrounding nodes, provides data support for abnormal propagation path simulation, performs topological anomaly propagation numerical simulation on the global topological structure model of the microgrid according to the abnormal node topological cascade effect data, simulates the propagation path of abnormal node failures in the microgrid, predicts the topological structure changes caused by abnormal node failures and the response of the microgrid system, provides a basis for local fracture analysis, performs topological local fracture analysis on the abnormal node propagation topological path to determine the local fracture area appearing in the abnormal topological structure, that is, the most severely affected area, obtains the abnormal topological local fracture area, quickly locates the area with the most severe abnormal impact, provides guidance for operation and maintenance personnel, and speeds up fault location and repair.

[0050] Preferably, the specific steps of step S53 are:

[0051] The propagation rate of the abnormal node topology impact value is calculated to obtain the abnormal topology propagation rate;

[0052] Based on the abnormal topology propagation rate, the abnormal topology propagation range is predicted for the abnormal node topology cascade effect data to obtain the abnormal topology prediction range;

[0053] According to the abnormal topology prediction range, the global topology structure model of the microgrid is numerically simulated for the propagation of topological anomalies, thereby generating the topological path for the propagation of abnormal nodes.

[0054] The present invention calculates the propagation rate of the abnormal node topology impact value, determines the propagation rate of the abnormality in the microgrid, obtains the abnormal topology propagation rate, evaluates the speed and range of the abnormal propagation, provides a basis for predicting the abnormal propagation range and path, predicts the abnormal topology propagation range of the abnormal node topology cascade effect data based on the abnormal topology propagation rate, predicts the propagation range of the abnormality in the microgrid, obtains the abnormal topology prediction range, identifies the area affected by the abnormality in advance, provides predictive support for early warning and response to the abnormality, performs topology abnormality propagation numerical simulation on the global topology structure model of the microgrid according to the abnormal topology prediction range, simulates the propagation path of the abnormality in the microgrid, understands the specific path of the abnormal propagation and the affected nodes, provides guidance for operation and maintenance personnel, and helps to accurately respond to abnormal situations.

[0055] Preferably, the specific steps of step S6 are:

[0056] Step S61: performing regional topology node risk assessment on the abnormal topology local fracture area, thereby obtaining a topology fracture regional node risk assessment value;

[0057] Step S62: making personalized alarm decisions for each node based on the risk assessment values ​​of the nodes in the topology fracture area, so as to generate a personalized alarm strategy for each abnormal node;

[0058] Step S63: Perform intelligent decision optimization on the personalized alarm strategy of each abnormal node, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

[0059] The present invention performs regional topology node risk assessment on the abnormal topology local fracture area, assesses the risk level of the nodes in the area, obtains the risk assessment value of the nodes in the topology fracture area, identifies which nodes face greater risks under abnormal conditions, provides a basis for alarm decision-making, performs personalized alarm decision-making for each node based on the risk assessment value of the nodes in the topology fracture area, generates a personalized alarm strategy for each abnormal node, distinguishes the risk levels of different nodes, takes targeted alarm measures, improves the accuracy and practicality of the alarm, performs intelligent decision-making optimization on the personalized alarm strategy of each abnormal node, constructs a topology abnormality alarm optimization model, and executes microgrid topology abnormality alarm operations. Through intelligent decision-making optimization, the execution efficiency and accuracy of the alarm strategy are optimized to ensure that the system can make appropriate responses in a timely manner under abnormal conditions, thereby ensuring the safe and stable operation of the microgrid system.

[0060] In this specification, a multi-converter-based power grid topology abnormality alarm data processing system is provided, which is used to execute the multi-converter-based power grid topology abnormality alarm data processing method as described above, including:

[0061] The feature map module is used to identify the microgrid converter nodes based on the microgrid converter cluster and obtain the microgrid historical converter node state parameters; perform deep characterization and correlation mining on the microgrid historical converter node state parameters to build a node dynamic monitoring feature map;

[0062] The global topology module is used to mine the topological connection logic of microgrid converter nodes one by one, perform global topology fitting, and build a global topology model of the microgrid;

[0063] The trajectory deviation module is used to quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory;

[0064] The abnormal node positioning module is used to perform regular node filtering based on the abnormal dynamic state feature trajectory and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes;

[0065] A topological cascade module is used to perform topological cascade effect analysis on abnormal topological feature nodes to generate abnormal node topological cascade effect data; perform topological local fracture analysis based on the abnormal node topological cascade effect data to obtain abnormal topological local fracture areas;

[0066] The abnormal alarm module is used to make personalized alarm decisions for each node in the abnormal topology local fracture area, build a topology abnormal alarm optimization model, and execute microgrid topology abnormal alarm operations.

[0067] The present invention identifies microgrid converter nodes and obtains historical state parameters, establishes a dynamic feature map of the nodes, and based on deep characterization association mining, better understands the association relationship between nodes, provides a basis for subsequent abnormal detection and positioning, and can establish a global topological structure model of the microgrid by mining the topological connection logic of each node and fitting the global topological structure. Through this module, the system accurately understands the connection between nodes, provides more in-depth background information for abnormal detection and positioning, uses the feature map to calculate the trajectory deviation of the global topological structure, identifies the abnormal dynamic state feature trajectory, timely discovers the abnormal change of the node state, and warns and handles the problem. Abnormal nodes are located based on the abnormal dynamic state feature trajectory, accurately identify abnormal nodes, quickly respond to and accurately handle abnormal situations, reduce the impact of faults on the entire microgrid system, analyze the cascading effect of abnormal topological feature nodes, predict the propagation range and impact of abnormalities in the microgrid, take measures to prevent abnormal spread as soon as possible, ensure the stability and safety of the microgrid, and build personalized alarm decisions and optimization models for each node, provide refined strategies for abnormal alarms, improve the accuracy and practicality of abnormal alarms, enable the system to respond to potential problems in a timely manner, and ensure the normal operation of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A schematic flow chart of the steps of a method for processing abnormal alarm data of a power grid topology based on a multi-converter according to the present invention;

[0069] Figure 2 Detailed implementation flow chart of step S1;

[0070] Figure 3 Detailed implementation flow chart of step S2;

[0071] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0072] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0073] The present application example provides a method and system for processing data of abnormal power grid topology alarm based on multiple converters. The execution subjects of the method and system for processing data of abnormal power grid topology alarm based on multiple converters include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of audio and image management system, information management system, and cloud data management system.

[0074] See also Figures 1 to 4 The present invention provides a method for processing abnormal alarm data of power grid topology based on multi-converter, comprising the following steps:

[0075] Step S1: Identify microgrid converter nodes based on microgrid converter clusters and obtain microgrid historical converter node state parameters; perform deep characterization correlation mining on microgrid historical converter node state parameters to construct node dynamic monitoring feature maps;

[0076] Step S2: mining the topological connection logic of each microgrid converter node one by one, fitting the global topological structure, and constructing a microgrid global topological structure model;

[0077] Step S3: Quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory;

[0078] Step S4: Perform normalized node filtering based on the abnormal dynamic state feature trajectory, and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes;

[0079] Step S5: performing a topological cascade effect analysis on the abnormal topological feature nodes to generate abnormal node topological cascade effect data; performing a topological local fracture analysis based on the abnormal node topological cascade effect data to obtain an abnormal topological local fracture area;

[0080] Step S6: Make personalized alarm decisions for each node in the abnormal topology local fracture area, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

[0081] The present invention identifies microgrid converter nodes and obtains historical state parameters, establishes a historical data foundation for microgrid nodes, provides a basis for subsequent analysis, conducts deep characterization association mining and constructs node dynamic monitoring feature maps to discover association patterns between nodes, lays a foundation for anomaly detection, mines the topological connection logic of each node and fits the global topological structure to reveal the connection relationship between microgrid nodes, constructs a microgrid global topological structure model, and constructs a global topological structure model to understand the structure of the entire microgrid system, providing a global perspective for subsequent anomaly detection. The node dynamic monitoring feature map is used to perform quantitative calculation of feature trajectory deviation on the global topological structure model, discovers abnormal dynamic state feature trajectories, helps identify potential problems, and obtains abnormal dynamic state feature trajectories to detect microgrid problems in advance. Abnormal conditions in the network provide early warning for subsequent processing, perform regular node filtering based on the abnormal dynamic state feature trajectory and accurately locate the abnormal nodes to accurately find the abnormal nodes, narrow the scope of the problem, and obtain abnormal topological feature nodes to help operation and maintenance personnel quickly locate the problem and reduce troubleshooting time. Perform topological cascade effect analysis on abnormal topological feature nodes to understand the impact of abnormal nodes on the entire microgrid system, help assess risks and prioritize nodes, perform topological local fracture analysis based on abnormal node topological cascade effect data to determine the abnormal topological local fracture area, further refine the problem scope, make personalized alarm decisions for abnormal topological local fracture areas, and build a topological abnormality alarm optimization model to achieve more accurate alarms and processing, and improve the stability and security of the microgrid system.

[0082] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for processing abnormal power grid topology alarm data based on a multi-converter of the present invention. In this example, the steps of the method for processing abnormal power grid topology alarm data based on a multi-converter include:

[0083] Step S1: Identify microgrid converter nodes based on microgrid converter clusters and obtain microgrid historical converter node state parameters; perform deep characterization correlation mining on microgrid historical converter node state parameters to construct node dynamic monitoring feature maps;

[0084] In this embodiment, a large number of converter devices, such as transformers and commutation devices, are deployed in the microgrid. The monitoring data of these converters are used to identify each converter node in the microgrid. By analyzing the real-time operating parameters of the converter devices, the actual working status of each node in the current microgrid is determined. The operating data of each converter node in the microgrid over a period of time in the past, including parameters such as voltage, current, power, and frequency, are collected and sorted. A historical status database of the microgrid converter nodes is constructed to lay the foundation for subsequent in-depth analysis. Deep learning and other algorithms are used to analyze and mine the historical status data, explore the potential correlation patterns and inherent laws between the state parameters of each converter node, extract key indicators that can reflect the dynamic working characteristics of the node, and integrate them into a node dynamic monitoring feature map based on the key feature indicators. The map contains the time series change trend of the state parameters of each node and the correlation between the nodes. The feature map provides a basis for subsequent abnormal warning and fault diagnosis.

[0085] Step S2: mining the topological connection logic of each microgrid converter node one by one, and fitting the global topological structure to build a microgrid global topological structure model;

[0086] In this embodiment, for each microgrid converter node, the connection mode and connection strength between it and the surrounding nodes are analyzed, and the topological association rules between the nodes, such as the connection mode, connection weight, etc., are extracted. The topological connection relationship of each node is integrated, and a global topological structure model of the microgrid is constructed by using graph theory modeling and other methods. The model can reflect the connection relationship between all nodes in the microgrid, as well as the overall network topological characteristics. The fitted global topological structure model is digitally represented to form a computer-readable data structure. The model contains comprehensive information such as node attributes (location, type, etc.) and line attributes (length, impedance, etc.), which can be used for subsequent simulation analysis and fault prediction of the global topological structure of the microgrid.

[0087] Step S3: Quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory;

[0088] In this embodiment, the node dynamic monitoring feature map is associated with the constructed global topology structure model, and the dynamic state characteristics of each node are associated with its position and connection relationship in the global topology. Based on the fused model, the dynamic state feature trajectory of each node is gradually tracked and analyzed, and those abnormal dynamic state feature trajectories that deviate from the normal mode are identified. For each abnormal dynamic state feature trajectory, the degree of deviation from the normal mode is calculated, and a quantitative evaluation is performed using a method based on indicators such as distance and angle. The greater the degree of deviation, the more abnormal the dynamic state of the node. The above analysis results are summarized to form abnormal dynamic state feature trajectories of each node in the microgrid. These abnormal trajectories reflect that abnormal dynamic changes have occurred in certain areas of the microgrid, providing an important basis for subsequent fault diagnosis and prevention.

[0089] Step S4: Perform normalized node filtering based on the abnormal dynamic state feature trajectory, and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes;

[0090] In this embodiment, the abnormal dynamic state characteristic trajectory obtained above is compared and analyzed with the global topological structure model of the microgrid. According to the dynamic characteristics of the abnormal trajectory, its propagation path and influence range are located on the topological structure model, and all converter nodes involved in the abnormal trajectory in the topological structure are analyzed. These nodes are further analyzed and filtered, and some conventional nodes are eliminated. Those abnormal nodes that are obviously related to the abnormal trajectory are retained. The above-mentioned abnormal nodes are deeply analyzed to extract their key features in the topological structure. These features include the connection mode of the node, the state change law, etc. These features are combined as a description of the abnormal topological feature nodes.

[0091] Step S5: performing a topological cascade effect analysis on the abnormal topological feature nodes to generate abnormal node topological cascade effect data; performing a topological local fracture analysis based on the abnormal node topological cascade effect data to obtain an abnormal topological local fracture area;

[0092] In this embodiment, for the abnormal topological feature nodes obtained above, their importance in the microgrid topology structure is analyzed, and the centrality indicators in graph theory (such as betweenness centrality, closeness centrality, etc.) are used to quantify the impact of abnormal nodes on the stability of the topological structure, and simulate the cascade propagation process of abnormal node impact in the microgrid topology structure. The propagation range and intensity of the topological cascade effect caused by the abnormal nodes are obtained, and the key areas where the microgrid topology structure breaks during the propagation of abnormal node impact are analyzed. According to the range and severity of the fractured area, the abnormal local fracture area of ​​the microgrid topology is determined.

[0093] Step S6: Make personalized alarm decisions for each node in the abnormal topology local fracture area, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

[0094] In this embodiment, for the abnormal topology local fracture area obtained above, the importance of each node in the topological structure is analyzed, and the topological risk of each node is quantified using network centrality indicators, such as betweenness centrality and closeness centrality, and the topological importance of the node is integrated with the severity of the fracture area where it is located to obtain the topological risk assessment value of the node. According to the topological risk assessment value of each node, a corresponding personalized alarm strategy is formulated. The alarm strategy includes alarm level, alarm mode, alarm object and the like. Differentiated alarm actions are taken for nodes with different risk levels. Machine learning algorithms, such as reinforcement learning and genetic algorithms, are used to optimize the above alarm strategies. The optimization goals include minimizing the false alarm rate, improving the accuracy of alarms, reducing the response time, etc. A topological abnormality alarm optimization model is constructed, which can adaptively formulate the optimal alarm strategy for each node, and the optimized alarm strategy is applied to the real-time operation monitoring of the microgrid, the real-time monitoring of the microgrid topology status, and differentiated abnormal alarms are executed according to the optimization model.

[0095] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0096] Step S11: identifying microgrid converter nodes based on the microgrid converter cluster, and obtaining microgrid historical converter node state parameters;

[0097] Step S12: performing multi-time point voltage load calculation on microgrid historical converter node state parameters to generate converter node voltage load data;

[0098] Step S13: performing time series load trend analysis on the converter node voltage load data to obtain a node load trend curve;

[0099] Step S14: performing temperature state identification on microgrid historical converter node state parameters to generate node temperature characteristics;

[0100] Step S15: performing temperature rise variation trend evolution on the node temperature characteristics, thereby obtaining node temperature rise variation characteristic data;

[0101] Step S16: Based on the node temperature rise change characteristic data, deep characterization and correlation mining are performed on the node load trend curve to construct a node dynamic monitoring characteristic map.

[0102] In this embodiment, the converter equipment in the microgrid is used as the basis for node identification. By analyzing the state data of the converter cluster, each converter node in the microgrid is identified, and the historical working state parameters of these nodes, such as voltage, current, power factor, etc., are collected. The actual load of each node is calculated using the voltage, current and other state parameters of the node. For each converter node, a voltage load data sequence in its historical period is generated. Time series analysis methods, such as moving average, ARIMA, etc., are applied to predict the trend of the node load data, obtain the change trend characteristics of the load of each node, and form a node load trend curve. The environmental monitoring parameters of the node, such as ambient temperature and humidity, are used to identify the actual temperature state of the node. For each converter node, the temperature change characteristics in the historical period are generated, and the time evolution trend of the node temperature characteristics is analyzed, focusing on the temperature change rate. The characteristic parameters such as the node temperature change rate are extracted to form the node temperature rise change characteristic data. The deep learning and other technologies are used to explore the intrinsic relationship between the node temperature change and the load change, extract this correlation feature, and construct a comprehensive characterization model of the node dynamic state. Based on the characterization model, a characteristic map of node dynamic monitoring is established to provide a basis for fault diagnosis.

[0103] In this embodiment, the specific steps of step S16 are:

[0104] Based on the node temperature rise variation characteristic data, a temperature-load correlation analysis is performed on the node load trend curve to generate temperature-load correlation data;

[0105] Perform multi-dimensional feature space mapping on the converter node voltage load data and node temperature rise change feature data to generate a node feature multi-dimensional space model;

[0106] Based on the temperature-load correlation data, the node dynamic feature drift analysis is performed on the node feature multidimensional space model to generate the node dynamic drift trajectory;

[0107] Conduct deep characterization and correlation mining on the dynamic drift trajectory of nodes and construct a node dynamic monitoring feature map.

[0108] In this embodiment, real-time temperature rise change data of the node is collected, such as temperature data collected by thermocouples or thermal imaging, and real-time power load data of the node is collected at the same time, such as load information monitored by current, voltage, etc., and time series correlation analysis is performed on the temperature data and load data to establish a temperature-load association curve, and key characteristic parameters such as temperature rise sensitivity and load saturation point are extracted from the association curve. Multidimensional monitoring data such as temperature, voltage, and current are mapped to a high-dimensional feature space, and the main dimensions of the feature space are extracted using methods such as principal component analysis or kernel principal component analysis. A multidimensional feature space model of the node is constructed to describe the comprehensive working status of the node. The temperature-load association data is used to analyze the evolution trend of the node feature space model over time, identify abnormal drift trajectories in the feature space, and indicate abnormal changes in the working status of the node. The abnormal drift trajectories are visualized to reveal the law of changes in the dynamic characteristics of the node. Deep learning and other algorithms are used to extract implicit key features from the dynamic drift trajectories of the node, and a correlation network map of the dynamic characteristics of the node is constructed to describe the intrinsic connection between the features.

[0109] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0110] Step S21: Calculate the spatial position of the microgrid converter node to obtain the node spatial position coordinates;

[0111] Step S22: performing node spatial distribution analysis based on the node spatial position coordinates to obtain node spatial distribution data;

[0112] Step S23: performing microgrid topology analysis on the node spatial distribution data to obtain a microgrid topology diagram;

[0113] Step S24: mining the topological connection logic of each microgrid converter node one by one, and extracting the topological association logic between nodes;

[0114] Step S25: performing global topological structure fitting on the microgrid topological structure diagram according to the topological association logic between nodes, and constructing a global topological structure model of the microgrid.

[0115] In this embodiment, network communication positioning technology, such as RFID, Wi-Fi positioning, etc., is used to obtain the specific position of each converter node in space, and the obtained node position information is converted into spatial coordinates in a three-dimensional coordinate system to describe the specific distribution of the node. The spatial coordinate data of the node is clustered to identify the node aggregation in different areas. According to the node distribution characteristics, key parameters such as the distribution density and distribution range of the node in space are extracted. The above analysis results are integrated to generate the spatial distribution data of the microgrid converter node. The node spatial position information is used to construct the connection relationship between the nodes in the microgrid. The graph theory analysis method is applied to generate the topological structure diagram of the microgrid according to the node position and connection status. For each node, the connection mode and connection strength with the surrounding nodes are analyzed, and the topological association rules between the nodes, such as the connection mode and connection weight, are extracted. The topological association logic of each node is integrated to form a global node topological association model of the microgrid. The topological association rules between the nodes are used to deeply optimize and correct the topological structure diagram. Based on the optimized topological structure, a global topological structure model of the microgrid is constructed, which accurately describes the connection relationship and transmission characteristics between the nodes in the microgrid.

[0116] In this embodiment, reference Figure 4 The above is a schematic flow chart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0117] Step S31: performing node dynamic state characteristic monitoring simulation on the microgrid global topology structure model to extract node dynamic state characteristic data;

[0118] Step S32: performing time series trajectory evolution on the node dynamic state feature data to generate a node dynamic state evolution trajectory;

[0119] Step S33: using the node dynamic monitoring characteristic map to perform quantitative calculation of characteristic trajectory deviation on the node dynamic state evolution trajectory to obtain dynamic characteristic trajectory deviation data;

[0120] Step S34: performing abnormal trajectory feature analysis on the dynamic feature trajectory deviation data, thereby obtaining an abnormal dynamic state feature trajectory.

[0121] In this embodiment, based on the constructed microgrid global topology model, the working state of each node is dynamically simulated. During the simulation process, the changes in dynamic parameters such as voltage, current, and power of the monitoring node are monitored, and the statistical characteristics of the above dynamic parameters, such as mean value, standard deviation, correlation coefficient, etc., are extracted as node dynamic state feature data. The node dynamic state feature data are arranged in time series to form a state feature evolution curve of each node. For each node, the evolution trajectory of its dynamic state characteristics in the time dimension is described, and the node dynamic state evolution trajectory is compared with the node dynamic monitoring feature map constructed previously. The degree of difference between the node dynamic state evolution trajectory and the normal feature model is quantified to obtain a trajectory deviation degree index. For the dynamic feature trajectory deviation data, abnormal trajectories with a deviation degree exceeding the normal range are identified, the characteristics of these abnormal trajectories are analyzed, and the key feature parameters of the abnormal node dynamic state are extracted.

[0122] In this embodiment, step S4 includes the following steps:

[0123] Step S41: dynamically positioning the global topology structure model of the microgrid based on the abnormal dynamic state characteristic trajectory to obtain abnormal trajectory positioning data;

[0124] Step S42: performing node identification on the abnormal trajectory positioning data, and marking all converter nodes on the abnormal trajectory;

[0125] Step S43: performing normalized node filtering on all converter nodes on the abnormal trajectory and accurately locating the abnormal nodes, thereby obtaining abnormal topological feature nodes.

[0126] In this embodiment, the abnormal dynamic state characteristic trajectory is superimposed and compared with the global topological structure model of the microgrid, and the abnormal operation trajectory is located on the topological structure model according to the dynamic change characteristics of the abnormal trajectory, and key information such as the spatial position and propagation path of the abnormal trajectory in the topological structure is extracted to form abnormal trajectory positioning data, and the propagation path of the abnormal trajectory in the topological structure is analyzed to identify all converter nodes passing along the way, mark these nodes, form a node list on the abnormal trajectory, further analyze and filter the node list on the abnormal trajectory, eliminate some conventional converter nodes, and retain those abnormal nodes that are obviously associated with the abnormal trajectory, and use the topological characteristic parameters of these abnormal nodes, such as connection mode, state change, etc., as the final abnormal topological features.

[0127] In this embodiment, step S5 includes the following steps:

[0128] Step S51: quantifying the topological stability impact of abnormal topological characteristic nodes to generate abnormal node topological impact values;

[0129] Step S52: performing a topological cascade effect analysis on the abnormal node topological impact value to generate abnormal node topological cascade effect data;

[0130] Step S53: performing a numerical simulation of topological anomaly propagation on the global topological structure model of the microgrid according to the abnormal node topological cascade effect data, thereby generating an abnormal node propagation topological path;

[0131] Step S54: Perform a topological local fracture analysis on the abnormal node propagation topological path to obtain an abnormal topological local fracture area.

[0132] In this embodiment, for the obtained abnormal topological feature nodes, their importance in the microgrid topological structure is analyzed, and the centrality indicators in graph theory (such as betweenness centrality, closeness centrality, etc.) are used to quantify the impact of abnormal nodes on the stability of the topological structure, and obtain the topological impact value of each abnormal node to characterize its abnormality. The degree of influence of the abnormal node topological impact on the surrounding neighboring nodes is analyzed, and the cascade propagation process of the abnormal node impact in the microgrid topological structure is simulated to obtain data such as the propagation range and intensity of the topological cascade effect caused by the abnormal node. The abnormal node topological cascade effect data is input into the global topological structure model of the microgrid, and the dynamic propagation process of the abnormal node impact in the microgrid topological structure is simulated to obtain the specific propagation path of the abnormal node impact in the topological structure. The key area where the microgrid topological structure is broken during the propagation of the abnormal node impact is analyzed. The structural fracture includes microgrid line disconnection and microgrid node failure. According to the range and severity of the fracture area, the abnormal fracture area of ​​the microgrid topology is determined.

[0133] In this embodiment, the specific steps of step S53 are:

[0134] The propagation rate of the abnormal node topology impact value is calculated to obtain the abnormal topology propagation rate;

[0135] Based on the abnormal topology propagation rate, the abnormal topology propagation range is predicted for the abnormal node topology cascade effect data to obtain the abnormal topology prediction range;

[0136] According to the abnormal topology prediction range, the global topology structure model of the microgrid is numerically simulated for the propagation of topological anomalies, thereby generating the topological path for the propagation of abnormal nodes.

[0137] In this embodiment, the diffusion centrality index in graph theory is used to analyze the propagation speed of the abnormal node topology impact in the microgrid topology structure. For each abnormal node, the diffusion rate of its topology impact in the network is calculated to obtain the abnormal topology propagation rate. The abnormal node topology cascade effect data obtained above is fused with the abnormal topology propagation rate to predict the final propagation range of the abnormal node topology impact in the microgrid topology structure to obtain the abnormal topology prediction range. The abnormal topology prediction range is used as input to perform dynamic simulation on the global topology structure model of the microgrid to simulate the propagation process of the abnormal node topology impact in the entire microgrid topology structure, and obtain the specific propagation path of the abnormal node topology impact in the microgrid.

[0138] In this embodiment, the specific steps of step S6 are:

[0139] Step S61: performing regional topology node risk assessment on the abnormal topology local fracture area, thereby obtaining a topology fracture regional node risk assessment value;

[0140] Step S62: making personalized alarm decisions for each node based on the risk assessment values ​​of the nodes in the topology fracture area, so as to generate a personalized alarm strategy for each abnormal node;

[0141] Step S63: Perform intelligent decision optimization on the personalized alarm strategy of each abnormal node, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

[0142] In this embodiment, for the abnormal topological local fracture area obtained above, the importance of each node in the topological structure is analyzed, and the topological importance of each node is quantified using network centrality indicators, such as betweenness centrality, closeness centrality, etc. The topological importance of the node is integrated with the severity of the fracture area where it is located to obtain the topological risk assessment value of the node. For each abnormal node, a corresponding personalized alarm strategy is formulated according to its topological risk assessment value. The alarm strategy includes alarm level, alarm method, alarm object, etc. Differentiated alarm actions are taken for nodes with different risk levels. Machine learning algorithms, such as reinforcement learning and genetic algorithms, are used to optimize the alarm strategy. The optimization goals include minimizing the false alarm rate, improving the accuracy of the alarm, reducing the response time, etc. A topological abnormality alarm optimization model is constructed, which can adaptively formulate the optimal alarm strategy for each node.

[0143] In this embodiment, a multi-converter-based power grid topology abnormality alarm data processing system is provided, which is used to execute the multi-converter-based power grid topology abnormality alarm data processing method as described above, including:

[0144] The feature map module is used to identify the microgrid converter nodes based on the microgrid converter cluster and obtain the microgrid historical converter node state parameters; perform deep characterization and correlation mining on the microgrid historical converter node state parameters to build a node dynamic monitoring feature map;

[0145] The global topology module is used to mine the topological connection logic of microgrid converter nodes one by one, perform global topology fitting, and build a global topology model of the microgrid;

[0146] The trajectory deviation module is used to quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory;

[0147] The abnormal node positioning module is used to perform regular node filtering based on the abnormal dynamic state feature trajectory and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes;

[0148] A topological cascade module is used to perform topological cascade effect analysis on abnormal topological feature nodes to generate abnormal node topological cascade effect data; perform topological local fracture analysis based on the abnormal node topological cascade effect data to obtain abnormal topological local fracture areas;

[0149] The abnormal alarm module is used to make personalized alarm decisions for each node in the abnormal topology local fracture area, build a topology abnormal alarm optimization model, and execute microgrid topology abnormal alarm operations.

[0150] The present invention identifies microgrid converter nodes and obtains historical state parameters, establishes a dynamic feature map of the nodes, and based on deep characterization association mining, better understands the association relationship between nodes, provides a basis for subsequent abnormal detection and positioning, and can establish a global topological structure model of the microgrid by mining the topological connection logic of each node and fitting the global topological structure. Through this module, the system accurately understands the connection between nodes, provides more in-depth background information for abnormal detection and positioning, uses the feature map to calculate the trajectory deviation of the global topological structure, identifies the abnormal dynamic state feature trajectory, timely discovers the abnormal change of the node state, and warns and handles the problem. Abnormal nodes are located based on the abnormal dynamic state feature trajectory, accurately identify abnormal nodes, quickly respond to and accurately handle abnormal situations, reduce the impact of faults on the entire microgrid system, analyze the cascading effect of abnormal topological feature nodes, predict the propagation range and impact of abnormalities in the microgrid, take measures to prevent abnormal spread as soon as possible, ensure the stability and safety of the microgrid, and build personalized alarm decisions and optimization models for each node, provide refined strategies for abnormal alarms, improve the accuracy and practicality of abnormal alarms, enable the system to respond to potential problems in a timely manner, and ensure the normal operation of the microgrid.

[0151] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0152] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for processing abnormal alarm data of power grid topology based on multi-converter, characterized in that: The following steps are involved: Step S1: Identify microgrid converter nodes based on microgrid converter clusters and obtain microgrid historical converter node state parameters; perform deep characterization correlation mining on microgrid historical converter node state parameters to construct node dynamic monitoring feature maps; Step S2: mining the topological connection logic of each microgrid converter node one by one, fitting the global topological structure, and building a global topological structure model of the microgrid; Step S3: Quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory; Step S4: Perform normalized node filtering based on the abnormal dynamic state feature trajectory, and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes; Step S5: performing a topological cascade effect analysis on the abnormal topological feature nodes to generate abnormal node topological cascade effect data; performing a topological local fracture analysis based on the abnormal node topological cascade effect data to obtain an abnormal topological local fracture area; Step S6: Perform node-by-node personalized alarm decisions for the abnormal topology local fracture area, and build a topology abnormality alarm optimization model to perform microgrid topology abnormality alarm operations; Among them, the specific steps of step S1 are: Step S11: identifying microgrid converter nodes based on the microgrid converter cluster, and obtaining microgrid historical converter node state parameters; Step S12: performing multi-time point voltage load calculation on microgrid historical converter node state parameters to generate converter node voltage load data; Step S13: performing time series load trend analysis on the converter node voltage load data to obtain a node load trend curve; Step S14: performing temperature state identification on microgrid historical converter node state parameters to generate node temperature characteristics; Step S15: performing temperature rise variation trend evolution on the node temperature characteristics, thereby obtaining node temperature rise variation characteristic data; Step S16: Based on the node temperature rise change characteristic data, deep characterization and correlation mining are performed on the node load trend curve to construct a node dynamic monitoring characteristic map.

2. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S16 are: Based on the node temperature rise variation characteristic data, a temperature-load correlation analysis is performed on the node load trend curve to generate temperature-load correlation data; Perform multi-dimensional feature space mapping on the converter node voltage load data and node temperature rise change feature data to generate a node feature multi-dimensional space model; Based on the temperature-load correlation data, the node dynamic feature drift analysis is performed on the node feature multidimensional space model to generate the node dynamic drift trajectory; Deep characterization and correlation mining are carried out on the dynamic drift trajectory of nodes to construct a node dynamic monitoring feature map.

3. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: Calculate the spatial position of the microgrid converter node to obtain the node spatial position coordinates; Step S22: performing node spatial distribution analysis based on the node spatial position coordinates to obtain node spatial distribution data; Step S23: performing microgrid topology analysis on the node spatial distribution data to obtain a microgrid topology diagram; Step S24: mining the topological connection logic of each microgrid converter node one by one, and extracting the topological association logic between nodes; Step S25: performing global topological structure fitting on the microgrid topological structure diagram according to the topological association logic between nodes, and constructing a global topological structure model of the microgrid.

4. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing node dynamic state characteristic monitoring simulation on the microgrid global topology structure model to extract node dynamic state characteristic data; Step S32: performing time series trajectory evolution on the node dynamic state feature data to generate a node dynamic state evolution trajectory; Step S33: using the node dynamic monitoring characteristic map to perform quantitative calculation of characteristic trajectory deviation on the node dynamic state evolution trajectory to obtain dynamic characteristic trajectory deviation data; Step S34: performing abnormal trajectory feature analysis on the dynamic feature trajectory deviation data, thereby obtaining an abnormal dynamic state feature trajectory.

5. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: dynamically positioning the global topology structure model of the microgrid based on the abnormal dynamic state characteristic trajectory to obtain abnormal trajectory positioning data; Step S42: performing node identification on the abnormal trajectory positioning data, and marking all converter nodes on the abnormal trajectory; Step S43: performing normal node filtering on all converter nodes on the abnormal trajectory and accurately locating the abnormal nodes, thereby obtaining abnormal topological feature nodes.

6. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: quantifying the topological stability impact of abnormal topological characteristic nodes to generate abnormal node topological impact values; Step S52: performing a topological cascade effect analysis on the abnormal node topological impact value to generate abnormal node topological cascade effect data; Step S53: performing a numerical simulation of topological anomaly propagation on the global topological structure model of the microgrid according to the abnormal node topological cascade effect data, thereby generating an abnormal node propagation topological path; Step S54: Perform a topological local fracture analysis on the abnormal node propagation topological path to obtain an abnormal topological local fracture area.

7. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 6, characterized in that: The specific steps of step S53 are: The propagation rate of the abnormal node topology impact value is calculated to obtain the abnormal topology propagation rate; Based on the abnormal topology propagation rate, the abnormal topology propagation range is predicted for the abnormal node topology cascade effect data to obtain the abnormal topology prediction range; According to the abnormal topology prediction range, the global topology structure model of the microgrid is numerically simulated for the propagation of topological anomalies, thereby generating the topological path for the propagation of abnormal nodes.

8. The method for processing abnormal power grid topology alarm data based on multi-converter according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing regional topology node risk assessment on the abnormal topology local fracture area, thereby obtaining a topology fracture regional node risk assessment value; Step S62: making personalized alarm decisions for each node based on the risk assessment values ​​of the nodes in the topology fracture area, so as to generate a personalized alarm strategy for each abnormal node; Step S63: Perform intelligent decision optimization on the personalized alarm strategy of each abnormal node, and build a topology abnormality alarm optimization model to execute the microgrid topology abnormality alarm operation.

9. A multi-converter-based power grid topology abnormality alarm data processing system, characterized in that: The method for processing abnormal power grid topology alarm data based on a multi-converter according to claim 1 comprises: The feature map module is used to identify the microgrid converter nodes based on the microgrid converter cluster and obtain the microgrid historical converter node state parameters; perform deep characterization and correlation mining on the microgrid historical converter node state parameters to build a node dynamic monitoring feature map; The global topology module is used to mine the topological connection logic of microgrid converter nodes one by one, perform global topology fitting, and build a global topology model of the microgrid; The trajectory deviation module is used to quantitatively calculate the characteristic trajectory deviation of the microgrid global topology structure model using the node dynamic monitoring characteristic map, so as to obtain the abnormal dynamic state characteristic trajectory; The abnormal node positioning module is used to perform regular node filtering based on the abnormal dynamic state feature trajectory and accurately locate the abnormal nodes, thereby obtaining abnormal topological feature nodes; A topological cascade module is used to perform topological cascade effect analysis on abnormal topological feature nodes to generate abnormal node topological cascade effect data; perform topological local fracture analysis based on the abnormal node topological cascade effect data to obtain abnormal topological local fracture areas; The abnormal alarm module is used to make personalized alarm decisions for each node in the abnormal topology local fracture area, build a topology abnormal alarm optimization model, and execute microgrid topology abnormal alarm operations.

Images